# caveat Published articles for caveat. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## The state of agent traffic in documentation (March 2026) DevFeed: [The state of agent traffic in documentation (March 2026)]() Original publisher: [Read original article]() Author: Han Wang Published: 2026-04-03T00:00:00Z Content type: article Language: en Sources: [Mintlify Blog]() Topics: [Documentation](), [ai-coding](), [Claude Code](), [cursor](), [Cloudflare](), [browsers]() Tags: [2026](), [agent](), [agents](), [ai](), [ai-agents](), [ai-coding](), [ai-coding-agents](), [ai-trends](), [analysis](), [browsers](), [caveat](), [claude-code](), [cloudflare](), [cursor](), [documentation](), [edge](), [firefox](), [http]() ### AI overview Mintlify analyzes 30 days of Cloudflare traffic across its documentation sites and finds that AI coding agents accounted for 45.3% of requests, nearly matching traditional browser traffic at 45.8%. Claude Code and Cursor generated 95.6% of identified AI-agent traffic, while OpenAI Codex could not be distinguished because it does not provide an identifying user-agent header. ### Source excerpt An analysis of 30 days of Mintlify documentation traffic, broken down by AI agents, browsers, and other sources. ## Resisting the Urge to Document Everything Everywhere DevFeed: [Resisting the Urge to Document Everything Everywhere]() Original publisher: [Read original article]() Author: lukaseder Published: 2025-02-28T10:39:07Z Content type: opinion Language: en Sources: [jOOQ]() Topics: [jooq](), [SQL](), [generics]() Tags: [caveat](), [documentation](), [generics](), [jooq](), [jooq-development](), [quirk](), [sql](), [stack-overflow](), [xyproblem]() ### AI overview The article argues that product teams should resist documenting every workaround alongside the feature that exposed a user's problem. Using jOOQ as an example, it explains that unrelated features can be difficult to discover even when one helps address a quirk or limitation in another. ### Source excerpt Every product manager knows this situation: This is such a common pattern, and while it's perfectly understandable for such a user to request this, it is so terribly wrong to give in to this user's request. Why is it wrong? The features are unrelated Most of the time, the two features X1 and X2 are ... Continue reading Resisting the Urge to Document Everything Everywhere -> ## Voice recognition DevFeed: [Voice recognition]() Original publisher: [Read original article]() Author: Adam Langley Published: 2023-07-29T00:00:00Z Content type: opinion Language: en Sources: [ImperialViolet]() Topics: [Whisper](), [iOS](), [Large Language Model]() Tags: [caveat](), [ios](), [llms](), [performance](), [script](), [transcription](), [voice](), [whisper]() ### AI overview The author compares iOS 17 voice recognition with earlier results and finds that, although performance improved over iOS 16, it still produces too many errors with technical terms. The article then reports that Whisper solved the practical voice-recognition problem, while noting that it can append fabricated sentences that are removable. ### Source excerpt Update: Evan let me know that Whisper solved the voice recognition problem. He has a wrapper that records from a microphone and prints the transcription here. Whisper is very impressive and the only caveat is that it sometimes inserts whole fabricated sentences at the end. The words always sort of make sense in context, but there were no sounds that could possibly have caused it. It's always at the very end in my experience, and it's no problem to remove it so, with that noted, you should ignore everything below because Whisper is a better answer. Last week's blog post was rather long, and had a greater than normal number of typos. (Thanks to people who pointed them out. I think I've fixed all the ones that were reported.) This was because I saw in reviews that iOS 17's voice recognition was supposed to be much improved, and I figured that I'd give it a try. I've always found iOS's recognition to be superior to Google Docs and I have an old iPad Pro that's good for betas. iOS's performance remains good and, yes, I think it's better than iOS 16. But it's still hardly at the level of "magic", especially when using technical terms. Here's a paragraph taken directly from the raw output of last week's post (I've highlighted errors with italics): It is integrated into the W3C credential management specification and so it is called via navigator . credentials . create and navigator .credentials. get. This document is about understanding the deeper structures that underpin web orphan rather than being a guy as to its details. So we will leave a great many details to the numerous guides to Web Oran that already exist on the web and instead focus on how structures from UF were carried over into Web orphan and updated. While it's nice that many of the words are there, with that density of errors doing all the corrections means that it's not clearly better than typing things out. However, the world is all aflutter about LLMs these days. Can they help? I wrote a script to chunk